Project Summary: Dynamic mapping of complex brain circuits by monitoring and modulating brain activity at a large scale will enhance our understanding of brain functions, such as sensation, thought, emotion, and action. This knowledge will ultimately help to better treat and prevent neurological disorders. Real-time interfacing with the brain also has the potential to enhance our perceptual, motor, and cognitive capabilities, as well as to restore sensory and motor functions lost through injury or disease. Despite decades of research and development of neurotechnologies for the brain, unfortunately monitoring and modulation of brain activity with high spatiotemporal resolution at a large scale is still one of the grand challenges in the 21st century. Current neural interfacing methods are either non-invasive, suffering from poor spatiotemporal resolution, or highly invasive (with higher specificity) that suffer from penetration into the brain parenchyma, causing scar tissue formation and long-term damage, while interfacing with only hundreds of neurons. Currently, neural dust is the only viable method that provides promise towards truly chronic high-spatiotemporal-resolution (?m, ms) electrophysiological recording and stimulation at a large scale. For neural dust to be successful, two fundamental technology innovations are required: 1) micron-scale free-floating independent implantable nodes with recording and stimulation capabilities, and 2) a subcranial miniaturized ultrasonic interrogation platform that can robustly and efficiently establish power and data links with the neural dust. While there have been significant efforts recently in developing micron-scale sensing and stimulation nodes, robust and efficient ultrasonic power and data transmission still remains the main obstacle in neural dust realization. Current state- of-the-art ultrasonic beamforming techniques are not practical due to: 1) post-implantation uncertainties, such as micro-motions and tissue interactions particularly in ambulatory subjects, since ultrasonic beamforming techniques require real-time knowledge of the implant?s location and propagation medium, particularly when used for interrogating free-floating micron-scale neural dust, and 2) utilizing rigid bulk piezoelectric transducers which cannot conform to uneven brain surfaces and are often too bulky for subcranial implantation. We propose a new ultrasonic interrogation (power/data) platform for neural dust, which virtually eliminates the aforementioned issues, with the following innovations: 1) Self-Image-Guided Ultrasonic (SIG-US) beamforming that can automatically adapt itself to the varying environment (micro-motion and tissue medium) without prior knowledge of the implant?s alignment/orientation and medium, leading to robust and efficient wireless power and data transmission, and 2) flexible thin-film ultrasonic transducer arrays that can conform to uneven brain surfaces and are suitable for subcranial implantation. To achieve this project, we have built a multidisciplinary team to address the electronics design and ultrasonic transducer design elements of this project.